The definitions of REFCOUNT_MAX and REFCOUNT_SATURATED are the same,
regardless of CONFIG_REFCOUNT_FULL, so consolidate them into a single
pair of definitions.
Signed-off-by: Will Deacon <will@kernel.org>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Hanjun Guo <guohanjun@huawei.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Elena Reshetova <elena.reshetova@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191121115902.2551-8-will@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Having the refcount saturation and warnings inline bloats the text,
despite the fact that these paths should never be executed in normal
operation.
Move the refcount saturation and warnings out of line to reduce the
image size when refcount_t checking is enabled. Relative to an x86_64
defconfig, the sizes reported by bloat-o-meter are:
# defconfig+REFCOUNT_FULL, inline saturation (i.e. before this patch)
Total: Before=14762076, After=14915442, chg +1.04%
# defconfig+REFCOUNT_FULL, out-of-line saturation (i.e. after this patch)
Total: Before=14762076, After=14835497, chg +0.50%
A side-effect of this change is that we now only get one warning per
refcount saturation type, rather than one per problematic call-site.
Signed-off-by: Will Deacon <will@kernel.org>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Hanjun Guo <guohanjun@huawei.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Elena Reshetova <elena.reshetova@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191121115902.2551-7-will@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Rewrite the generic REFCOUNT_FULL implementation so that the saturation
point is moved to INT_MIN / 2. This allows us to defer the sanity checks
until after the atomic operation, which removes many uses of cmpxchg()
in favour of atomic_fetch_{add,sub}().
Some crude perf results obtained from lkdtm show substantially less
overhead, despite the checking:
$ perf stat -r 3 -B -- echo {ATOMIC,REFCOUNT}_TIMING >/sys/kernel/debug/provoke-crash/DIRECT
# arm64
ATOMIC_TIMING: 46.50451 +- 0.00134 seconds time elapsed ( +- 0.00% )
REFCOUNT_TIMING (REFCOUNT_FULL, mainline): 77.57522 +- 0.00982 seconds time elapsed ( +- 0.01% )
REFCOUNT_TIMING (REFCOUNT_FULL, this series): 48.7181 +- 0.0256 seconds time elapsed ( +- 0.05% )
# x86
ATOMIC_TIMING: 31.6225 +- 0.0776 seconds time elapsed ( +- 0.25% )
REFCOUNT_TIMING (!REFCOUNT_FULL, mainline/x86 asm): 31.6689 +- 0.0901 seconds time elapsed ( +- 0.28% )
REFCOUNT_TIMING (REFCOUNT_FULL, mainline): 53.203 +- 0.138 seconds time elapsed ( +- 0.26% )
REFCOUNT_TIMING (REFCOUNT_FULL, this series): 31.7408 +- 0.0486 seconds time elapsed ( +- 0.15% )
Signed-off-by: Will Deacon <will@kernel.org>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Hanjun Guo <guohanjun@huawei.com>
Tested-by: Jan Glauber <jglauber@marvell.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Elena Reshetova <elena.reshetova@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191121115902.2551-6-will@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In an effort to improve performance of the REFCOUNT_FULL implementation,
move the bulk of its functions into linux/refcount.h. This allows them
to be inlined in the same way as if they had been provided via
CONFIG_ARCH_HAS_REFCOUNT.
Signed-off-by: Will Deacon <will@kernel.org>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Hanjun Guo <guohanjun@huawei.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Elena Reshetova <elena.reshetova@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191121115902.2551-5-will@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The full-fat refcount implementation is exposed via a set of functions
suffixed with "_checked()", the idea being that code can choose to use
the more expensive, yet more secure implementation on a case-by-case
basis.
In reality, this hasn't happened, so with a grand total of zero users,
let's remove the checked variants for now by simply dropping the suffix
and predicating the out-of-line functions on CONFIG_REFCOUNT_FULL=y.
Signed-off-by: Will Deacon <will@kernel.org>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Hanjun Guo <guohanjun@huawei.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Elena Reshetova <elena.reshetova@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191121115902.2551-4-will@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
In preparation for changing the saturation point of REFCOUNT_FULL to
INT_MIN/2, change the type of integer operands passed into the API
from 'unsigned int' to 'int' so that we can avoid casting during
comparisons when we don't want to fall foul of C integral conversion
rules for signed and unsigned types.
Since the kernel is compiled with '-fno-strict-overflow', we don't need
to worry about the UB introduced by signed overflow here. Furthermore,
we're already making heavy use of the atomic_t API, which operates
exclusively on signed types.
Signed-off-by: Will Deacon <will@kernel.org>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Hanjun Guo <guohanjun@huawei.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Elena Reshetova <elena.reshetova@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191121115902.2551-3-will@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The REFCOUNT_FULL implementation uses a different saturation point than
the x86 implementation, which means that the shared refcount code in
lib/refcount.c (e.g. refcount_dec_not_one()) needs to be aware of the
difference.
Rather than duplicate the definitions from the lkdtm driver, instead
move them into <linux/refcount.h> and update all references accordingly.
Signed-off-by: Will Deacon <will@kernel.org>
Reviewed-by: Ard Biesheuvel <ardb@kernel.org>
Reviewed-by: Kees Cook <keescook@chromium.org>
Tested-by: Hanjun Guo <guohanjun@huawei.com>
Cc: Ard Biesheuvel <ard.biesheuvel@linaro.org>
Cc: Elena Reshetova <elena.reshetova@intel.com>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191121115902.2551-2-will@kernel.org
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Oleg provided the following test case:
int main(void)
{
struct sched_param sp = {};
sp.sched_priority = 2;
assert(sched_setscheduler(0, SCHED_FIFO, &sp) == 0);
int lock = vfork();
if (!lock) {
sp.sched_priority = 1;
assert(sched_setscheduler(0, SCHED_FIFO, &sp) == 0);
_exit(0);
}
syscall(__NR_futex, &lock, FUTEX_LOCK_PI, 0,0,0);
return 0;
}
This creates an unkillable RT process spinning in futex_lock_pi() on a UP
machine or if the process is affine to a single CPU. The reason is:
parent child
set FIFO prio 2
vfork() -> set FIFO prio 1
implies wait_for_child() sched_setscheduler(...)
exit()
do_exit()
....
mm_release()
tsk->futex_state = FUTEX_STATE_EXITING;
exit_futex(); (NOOP in this case)
complete() --> wakes parent
sys_futex()
loop infinite because
tsk->futex_state == FUTEX_STATE_EXITING
The same problem can happen just by regular preemption as well:
task holds futex
...
do_exit()
tsk->futex_state = FUTEX_STATE_EXITING;
--> preemption (unrelated wakeup of some other higher prio task, e.g. timer)
switch_to(other_task)
return to user
sys_futex()
loop infinite as above
Just for the fun of it the futex exit cleanup could trigger the wakeup
itself before the task sets its futex state to DEAD.
To cure this, the handling of the exiting owner is changed so:
- A refcount is held on the task
- The task pointer is stored in a caller visible location
- The caller drops all locks (hash bucket, mmap_sem) and blocks
on task::futex_exit_mutex. When the mutex is acquired then
the exiting task has completed the cleanup and the state
is consistent and can be reevaluated.
This is not a pretty solution, but there is no choice other than returning
an error code to user space, which would break the state consistency
guarantee and open another can of problems including regressions.
For stable backports the preparatory commits ac31c7ff86 .. ba31c1a485
are required as well, but for anything older than 5.3.y the backports are
going to be provided when this hits mainline as the other dependencies for
those kernels are definitely not stable material.
Fixes: 778e9a9c3e ("pi-futex: fix exit races and locking problems")
Reported-by: Oleg Nesterov <oleg@redhat.com>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Stable Team <stable@vger.kernel.org>
Link: https://lkml.kernel.org/r/20191106224557.041676471@linutronix.de
attach_to_pi_owner() returns -EAGAIN for various cases:
- Owner task is exiting
- Futex value has changed
The caller drops the held locks (hash bucket, mmap_sem) and retries the
operation. In case of the owner task exiting this can result in a live
lock.
As a preparatory step for seperating those cases, provide a distinct return
value (EBUSY) for the owner exiting case.
No functional change.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.935606117@linutronix.de
The mutex will be used in subsequent changes to replace the busy looping of
a waiter when the futex owner is currently executing the exit cleanup to
prevent a potential live lock.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.845798895@linutronix.de
exec() attempts to handle potentially held futexes gracefully by running
the futex exit handling code like exit() does.
The current implementation has no protection against concurrent incoming
waiters. The reason is that the futex state cannot be set to
FUTEX_STATE_DEAD after the cleanup because the task struct is still active
and just about to execute the new binary.
While its arguably buggy when a task holds a futex over exec(), for
consistency sake the state handling can at least cover the actual futex
exit cleanup section. This provides state consistency protection accross
the cleanup. As the futex state of the task becomes FUTEX_STATE_OK after the
cleanup has been finished, this cannot prevent subsequent attempts to
attach to the task in case that the cleanup was not successfull in mopping
up all leftovers.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.753355618@linutronix.de
Instead of having a smp_mb() and an empty lock/unlock of task::pi_lock move
the state setting into to the lock section.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.645603214@linutronix.de
Instead of relying on PF_EXITING use an explicit state for the futex exit
and set it in the futex exit function. This moves the smp barrier and the
lock/unlock serialization into the futex code.
As with the DEAD state this is restricted to the exit path as exec
continues to use the same task struct.
This allows to simplify that logic in a next step.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.539409004@linutronix.de
Setting task::futex_state in do_exit() is rather arbitrarily placed for no
reason. Move it into the futex code.
Note, this is only done for the exit cleanup as the exec cleanup cannot set
the state to FUTEX_STATE_DEAD because the task struct is still in active
use.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.439511191@linutronix.de
To allow separate handling of the futex exit state in the futex exit code
for exit and exec, split futex_mm_release() into two functions and invoke
them from the corresponding exit/exec_mm_release() callsites.
Preparatory only, no functional change.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.332094221@linutronix.de
mm_release() contains the futex exit handling. mm_release() is called from
do_exit()->exit_mm() and from exec()->exec_mm().
In the exit_mm() case PF_EXITING and the futex state is updated. In the
exec_mm() case these states are not touched.
As the futex exit code needs further protections against exit races, this
needs to be split into two functions.
Preparatory only, no functional change.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.240518241@linutronix.de
The futex exit handling relies on PF_ flags. That's suboptimal as it
requires a smp_mb() and an ugly lock/unlock of the exiting tasks pi_lock in
the middle of do_exit() to enforce the observability of PF_EXITING in the
futex code.
Add a futex_state member to task_struct and convert the PF_EXITPIDONE logic
over to the new state. The PF_EXITING dependency will be cleaned up in a
later step.
This prepares for handling various futex exit issues later.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.149449274@linutronix.de
The futex exit handling is #ifdeffed into mm_release() which is not pretty
to begin with. But upcoming changes to address futex exit races need to add
more functionality to this exit code.
Split it out into a function, move it into futex code and make the various
futex exit functions static.
Preparatory only and no functional change.
Folded build fix from Borislav.
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Link: https://lkml.kernel.org/r/20191106224556.049705556@linutronix.de
Robust futexes utilize the robust_list mechanism to allow the kernel to
release futexes which are held when a task exits. The exit can be voluntary
or caused by a signal or fault. This prevents that waiters block forever.
The futex operations in user space store a pointer to the futex they are
either locking or unlocking in the op_pending member of the per task robust
list.
After a lock operation has succeeded the futex is queued in the robust list
linked list and the op_pending pointer is cleared.
After an unlock operation has succeeded the futex is removed from the
robust list linked list and the op_pending pointer is cleared.
The robust list exit code checks for the pending operation and any futex
which is queued in the linked list. It carefully checks whether the futex
value is the TID of the exiting task. If so, it sets the OWNER_DIED bit and
tries to wake up a potential waiter.
This is race free for the lock operation but unlock has two race scenarios
where waiters might not be woken up. These issues can be observed with
regular robust pthread mutexes. PI aware pthread mutexes are not affected.
(1) Unlocking task is killed after unlocking the futex value in user space
before being able to wake a waiter.
pthread_mutex_unlock()
|
V
atomic_exchange_rel (&mutex->__data.__lock, 0)
<------------------------killed
lll_futex_wake () |
|
|(__lock = 0)
|(enter kernel)
|
V
do_exit()
exit_mm()
mm_release()
exit_robust_list()
handle_futex_death()
|
|(__lock = 0)
|(uval = 0)
|
V
if ((uval & FUTEX_TID_MASK) != task_pid_vnr(curr))
return 0;
The sanity check which ensures that the user space futex is owned by
the exiting task prevents the wakeup of waiters which in consequence
block infinitely.
(2) Waiting task is killed after a wakeup and before it can acquire the
futex in user space.
OWNER WAITER
futex_wait()
pthread_mutex_unlock() |
| |
|(__lock = 0) |
| |
V |
futex_wake() ------------> wakeup()
|
|(return to userspace)
|(__lock = 0)
|
V
oldval = mutex->__data.__lock
<-----------------killed
atomic_compare_and_exchange_val_acq (&mutex->__data.__lock, |
id | assume_other_futex_waiters, 0) |
|
|
(enter kernel)|
|
V
do_exit()
|
|
V
handle_futex_death()
|
|(__lock = 0)
|(uval = 0)
|
V
if ((uval & FUTEX_TID_MASK) != task_pid_vnr(curr))
return 0;
The sanity check which ensures that the user space futex is owned
by the exiting task prevents the wakeup of waiters, which seems to
be correct as the exiting task does not own the futex value, but
the consequence is that other waiters wont be woken up and block
infinitely.
In both scenarios the following conditions are true:
- task->robust_list->list_op_pending != NULL
- user space futex value == 0
- Regular futex (not PI)
If these conditions are met then it is reasonably safe to wake up a
potential waiter in order to prevent the above problems.
As this might be a false positive it can cause spurious wakeups, but the
waiter side has to handle other types of unrelated wakeups, e.g. signals
gracefully anyway. So such a spurious wakeup will not affect the
correctness of these operations.
This workaround must not touch the user space futex value and cannot set
the OWNER_DIED bit because the lock value is 0, i.e. uncontended. Setting
OWNER_DIED in this case would result in inconsistent state and subsequently
in malfunction of the owner died handling in user space.
The rest of the user space state is still consistent as no other task can
observe the list_op_pending entry in the exiting tasks robust list.
The eventually woken up waiter will observe the uncontended lock value and
take it over.
[ tglx: Massaged changelog and comment. Made the return explicit and not
depend on the subsequent check and added constants to hand into
handle_futex_death() instead of plain numbers. Fixed a few coding
style issues. ]
Fixes: 0771dfefc9 ("[PATCH] lightweight robust futexes: core")
Signed-off-by: Yang Tao <yang.tao172@zte.com.cn>
Signed-off-by: Yi Wang <wang.yi59@zte.com.cn>
Signed-off-by: Thomas Gleixner <tglx@linutronix.de>
Reviewed-by: Ingo Molnar <mingo@kernel.org>
Acked-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: stable@vger.kernel.org
Link: https://lkml.kernel.org/r/1573010582-35297-1-git-send-email-wang.yi59@zte.com.cn
Link: https://lkml.kernel.org/r/20191106224555.943191378@linutronix.de
This changes "to the list" to "from the list" and also deletes the
obsolete comment about the "@nested" argument.
The "nested" argument was removed in this commit, earlier this year:
5facae4f35 ("locking/lockdep: Remove unused @nested argument from lock_release()").
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Andrew Morton <akpm@linux-foundation.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Paul E. McKenney <paulmck@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Will Deacon <will.deacon@arm.com>
Cc: Will Deacon <will@kernel.org>
Link: https://lkml.kernel.org/r/20191104091252.GA31509@mwanda
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Add warning checks if mutex_trylock() or mutex_unlock() are used in
IRQ contexts, under CONFIG_DEBUG_MUTEXES=y.
While the mutex rules and semantics are explicitly documented, this allows
to expose any abusers and robustifies the whole thing.
While trylock and unlock are non-blocking, calling from IRQ context
is still forbidden (lock must be within the same context as unlock).
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dave@stgolabs.net
Link: https://lkml.kernel.org/r/20191025033634.3330-1-dave@stgolabs.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
Since the original comment, we have moved to do the task
reference counting explicitly along with wake_q_add_safe().
Drop the now incorrect comment.
Signed-off-by: Davidlohr Bueso <dbueso@suse.de>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: dave@stgolabs.net
Link: https://lkml.kernel.org/r/20191023033450.6445-1-dave@stgolabs.net
Signed-off-by: Ingo Molnar <mingo@kernel.org>
The check_preemption_disabled() function uses cpumask_equal() to see
if the task is bounded to the current CPU only. cpumask_equal() calls
memcmp() to do the comparison. As x86 doesn't have __HAVE_ARCH_MEMCMP,
the slow memcmp() function in lib/string.c is used.
On a RT kernel that call check_preemption_disabled() very frequently,
below is the perf-record output of a certain microbenchmark:
42.75% 2.45% testpmd [kernel.kallsyms] [k] check_preemption_disabled
40.01% 39.97% testpmd [kernel.kallsyms] [k] memcmp
We should avoid calling memcmp() in performance critical path. So the
cpumask_equal() call is now replaced with an equivalent simpler check.
Signed-off-by: Waiman Long <longman@redhat.com>
Signed-off-by: Peter Zijlstra (Intel) <peterz@infradead.org>
Reviewed-by: Juri Lelli <juri.lelli@redhat.com>
Acked-by: Sebastian Andrzej Siewior <bigeasy@linutronix.de>
Cc: Linus Torvalds <torvalds@linux-foundation.org>
Cc: Masami Hiramatsu <mhiramat@kernel.org>
Cc: Peter Zijlstra <peterz@infradead.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Link: https://lkml.kernel.org/r/20191003203608.21881-1-longman@redhat.com
Signed-off-by: Ingo Molnar <mingo@kernel.org>
-----BEGIN PGP SIGNATURE-----
iHUEABYIAB0WIQQUwxxKyE5l/npt8ARiEGxRG/Sl2wUCXZzy3AAKCRBiEGxRG/Sl
2wFjAP9BFsZMqiz8wTxlvn/vuVXg8V1TAbJzwn0rJJKPJsggnQD9HFBKJ3Vq995R
C02zqXE+8wJlqCGRK4pmJey5KamjLQo=
=JF8E
-----END PGP SIGNATURE-----
Merge tag 'led-fixes-for-5.4-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds
Pull LED fixes from Jacek Anaszewski:
- fix a leftover from earlier stage of development in the documentation
of recently added led_compose_name() and fix old mistake in the
documentation of led_set_brightness_sync() parameter name.
- MAINTAINERS: add pointer to Pavel Machek's linux-leds.git tree.
Pavel is going to take over LED tree maintainership from myself.
* tag 'led-fixes-for-5.4-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/j.anaszewski/linux-leds:
Add my linux-leds branch to MAINTAINERS
leds: core: Fix leds.h structure documentation
Add pointer to my git tree to MAINTAINERS. I'd like to maintain
linux-leds for-next branch for 5.5.
Signed-off-by: Pavel Machek <pavel@ucw.cz>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
Update the leds.h structure documentation to define the
correct arguments.
Signed-off-by: Dan Murphy <dmurphy@ti.com>
Signed-off-by: Jacek Anaszewski <jacek.anaszewski@gmail.com>
- Don't clear FLAG_IS_OUT when emulating open drain/source in
gpiolib.
- Fix up the usage of nonexclusive GPIO descriptors from device
trees.
- Fix the incorrect IEC offset when toggling trigger edge in
the Spreadtrum driver.
- Use the correct unit for debounce settings in the MAX77620
driver.
-----BEGIN PGP SIGNATURE-----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=J0x0
-----END PGP SIGNATURE-----
Merge tag 'gpio-v5.4-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio
Pull GPIO fixes from Linus Walleij:
- don't clear FLAG_IS_OUT when emulating open drain/source in gpiolib
- fix up the usage of nonexclusive GPIO descriptors from device trees
- fix the incorrect IEC offset when toggling trigger edge in the
Spreadtrum driver
- use the correct unit for debounce settings in the MAX77620 driver
* tag 'gpio-v5.4-2' of git://git.kernel.org/pub/scm/linux/kernel/git/linusw/linux-gpio:
gpio: max77620: Use correct unit for debounce times
gpio: eic: sprd: Fix the incorrect EIC offset when toggling
gpio: fix getting nonexclusive gpiods from DT
gpiolib: don't clear FLAG_IS_OUT when emulating open-drain/open-source
-----BEGIN PGP SIGNATURE-----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=btWt
-----END PGP SIGNATURE-----
Merge tag 'selinux-pr-20191007' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux
Pull selinuxfix from Paul Moore:
"One patch to ensure we don't copy bad memory up into userspace"
* tag 'selinux-pr-20191007' of git://git.kernel.org/pub/scm/linux/kernel/git/pcmoore/selinux:
selinux: fix context string corruption in convert_context()
This Kselftest update for Linux 5.4-rc3 consists fixes for existing
tests and the framework. Cristian Marussi's patches add ability to
skip targets (tests) and exclude tests that didn't build from run-list.
These patches improve the Kselftest results. Ability to skip targets
helps avoid running tests that aren't supported in certain environments.
As an example, bpf tests from mainline aren't supported on stable kernels
and have dependency on bleeding edge llvm. Being able to skip bpf on
systems that can't meet this llvm dependency will be helpful.
Kselftest can be built and installed from the main Makefile. This change
help simplify Kselftest use-cases which addresses request from users.
Kees Cook added per test timeout support to limit individual test run-time.
-----BEGIN PGP SIGNATURE-----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=vdrT
-----END PGP SIGNATURE-----
Merge tag 'linux-kselftest-5.4-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest
Pull Kselftest fixes from Shuah Khan:
"Fixes for existing tests and the framework.
Cristian Marussi's patches add the ability to skip targets (tests) and
exclude tests that didn't build from run-list. These patches improve
the Kselftest results. Ability to skip targets helps avoid running
tests that aren't supported in certain environments. As an example,
bpf tests from mainline aren't supported on stable kernels and have
dependency on bleeding edge llvm. Being able to skip bpf on systems
that can't meet this llvm dependency will be helpful.
Kselftest can be built and installed from the main Makefile. This
change help simplify Kselftest use-cases which addresses request from
users.
Kees Cook added per test timeout support to limit individual test
run-time"
* tag 'linux-kselftest-5.4-rc3' of git://git.kernel.org/pub/scm/linux/kernel/git/shuah/linux-kselftest:
selftests: watchdog: Add command line option to show watchdog_info
selftests: watchdog: Validate optional file argument
selftests/kselftest/runner.sh: Add 45 second timeout per test
kselftest: exclude failed TARGETS from runlist
kselftest: add capability to skip chosen TARGETS
selftests: Add kselftest-all and kselftest-install targets
Merge misc fixes from Andrew Morton:
"The usual shower of hotfixes.
Chris's memcg patches aren't actually fixes - they're mature but a few
niggling review issues were late to arrive.
The ocfs2 fixes are quite old - those took some time to get reviewer
attention.
Subsystems affected by this patch series: ocfs2, hotfixes, mm/memcg,
mm/slab-generic"
* emailed patches from Andrew Morton <akpm@linux-foundation.org>:
mm, sl[aou]b: guarantee natural alignment for kmalloc(power-of-two)
mm, sl[ou]b: improve memory accounting
mm, memcg: make scan aggression always exclude protection
mm, memcg: make memory.emin the baseline for utilisation determination
mm, memcg: proportional memory.{low,min} reclaim
mm/vmpressure.c: fix a signedness bug in vmpressure_register_event()
mm/page_alloc.c: fix a crash in free_pages_prepare()
mm/z3fold.c: claim page in the beginning of free
kernel/sysctl.c: do not override max_threads provided by userspace
memcg: only record foreign writebacks with dirty pages when memcg is not disabled
mm: fix -Wmissing-prototypes warnings
writeback: fix use-after-free in finish_writeback_work()
mm/memremap: drop unused SECTION_SIZE and SECTION_MASK
panic: ensure preemption is disabled during panic()
fs: ocfs2: fix a possible null-pointer dereference in ocfs2_info_scan_inode_alloc()
fs: ocfs2: fix a possible null-pointer dereference in ocfs2_write_end_nolock()
fs: ocfs2: fix possible null-pointer dereferences in ocfs2_xa_prepare_entry()
ocfs2: clear zero in unaligned direct IO
In most configurations, kmalloc() happens to return naturally aligned
(i.e. aligned to the block size itself) blocks for power of two sizes.
That means some kmalloc() users might unknowingly rely on that
alignment, until stuff breaks when the kernel is built with e.g.
CONFIG_SLUB_DEBUG or CONFIG_SLOB, and blocks stop being aligned. Then
developers have to devise workaround such as own kmem caches with
specified alignment [1], which is not always practical, as recently
evidenced in [2].
The topic has been discussed at LSF/MM 2019 [3]. Adding a
'kmalloc_aligned()' variant would not help with code unknowingly relying
on the implicit alignment. For slab implementations it would either
require creating more kmalloc caches, or allocate a larger size and only
give back part of it. That would be wasteful, especially with a generic
alignment parameter (in contrast with a fixed alignment to size).
Ideally we should provide to mm users what they need without difficult
workarounds or own reimplementations, so let's make the kmalloc()
alignment to size explicitly guaranteed for power-of-two sizes under all
configurations. What this means for the three available allocators?
* SLAB object layout happens to be mostly unchanged by the patch. The
implicitly provided alignment could be compromised with
CONFIG_DEBUG_SLAB due to redzoning, however SLAB disables redzoning for
caches with alignment larger than unsigned long long. Practically on at
least x86 this includes kmalloc caches as they use cache line alignment,
which is larger than that. Still, this patch ensures alignment on all
arches and cache sizes.
* SLUB layout is also unchanged unless redzoning is enabled through
CONFIG_SLUB_DEBUG and boot parameter for the particular kmalloc cache.
With this patch, explicit alignment is guaranteed with redzoning as
well. This will result in more memory being wasted, but that should be
acceptable in a debugging scenario.
* SLOB has no implicit alignment so this patch adds it explicitly for
kmalloc(). The potential downside is increased fragmentation. While
pathological allocation scenarios are certainly possible, in my testing,
after booting a x86_64 kernel+userspace with virtme, around 16MB memory
was consumed by slab pages both before and after the patch, with
difference in the noise.
[1] https://lore.kernel.org/linux-btrfs/c3157c8e8e0e7588312b40c853f65c02fe6c957a.1566399731.git.christophe.leroy@c-s.fr/
[2] https://lore.kernel.org/linux-fsdevel/20190225040904.5557-1-ming.lei@redhat.com/
[3] https://lwn.net/Articles/787740/
[akpm@linux-foundation.org: documentation fixlet, per Matthew]
Link: http://lkml.kernel.org/r/20190826111627.7505-3-vbabka@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Reviewed-by: Matthew Wilcox (Oracle) <willy@infradead.org>
Acked-by: Michal Hocko <mhocko@suse.com>
Acked-by: Kirill A. Shutemov <kirill.shutemov@linux.intel.com>
Acked-by: Christoph Hellwig <hch@lst.de>
Cc: David Sterba <dsterba@suse.cz>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: "Darrick J . Wong" <darrick.wong@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Patch series "guarantee natural alignment for kmalloc()", v2.
This patch (of 2):
SLOB currently doesn't account its pages at all, so in /proc/meminfo the
Slab field shows zero. Modifying a counter on page allocation and
freeing should be acceptable even for the small system scenarios SLOB is
intended for. Since reclaimable caches are not separated in SLOB,
account everything as unreclaimable.
SLUB currently doesn't account kmalloc() and kmalloc_node() allocations
larger than order-1 page, that are passed directly to the page
allocator. As they also don't appear in /proc/slabinfo, it might look
like a memory leak. For consistency, account them as well. (SLAB
doesn't actually use page allocator directly, so no change there).
Ideally SLOB and SLUB would be handled in separate patches, but due to
the shared kmalloc_order() function and different kfree()
implementations, it's easier to patch both at once to prevent
inconsistencies.
Link: http://lkml.kernel.org/r/20190826111627.7505-2-vbabka@suse.cz
Signed-off-by: Vlastimil Babka <vbabka@suse.cz>
Cc: Christoph Lameter <cl@linux.com>
Cc: Pekka Enberg <penberg@kernel.org>
Cc: David Rientjes <rientjes@google.com>
Cc: Ming Lei <ming.lei@redhat.com>
Cc: Dave Chinner <david@fromorbit.com>
Cc: Matthew Wilcox <willy@infradead.org>
Cc: "Darrick J . Wong" <darrick.wong@oracle.com>
Cc: Christoph Hellwig <hch@lst.de>
Cc: James Bottomley <James.Bottomley@HansenPartnership.com>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Joonsoo Kim <iamjoonsoo.kim@lge.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
This patch is an incremental improvement on the existing
memory.{low,min} relative reclaim work to base its scan pressure
calculations on how much protection is available compared to the current
usage, rather than how much the current usage is over some protection
threshold.
This change doesn't change the experience for the user in the normal
case too much. One benefit is that it replaces the (somewhat arbitrary)
100% cutoff with an indefinite slope, which makes it easier to ballpark
a memory.low value.
As well as this, the old methodology doesn't quite apply generically to
machines with varying amounts of physical memory. Let's say we have a
top level cgroup, workload.slice, and another top level cgroup,
system-management.slice. We want to roughly give 12G to
system-management.slice, so on a 32GB machine we set memory.low to 20GB
in workload.slice, and on a 64GB machine we set memory.low to 52GB.
However, because these are relative amounts to the total machine size,
while the amount of memory we want to generally be willing to yield to
system.slice is absolute (12G), we end up putting more pressure on
system.slice just because we have a larger machine and a larger workload
to fill it, which seems fairly unintuitive. With this new behaviour, we
don't end up with this unintended side effect.
Previously the way that memory.low protection works is that if you are
50% over a certain baseline, you get 50% of your normal scan pressure.
This is certainly better than the previous cliff-edge behaviour, but it
can be improved even further by always considering memory under the
currently enforced protection threshold to be out of bounds. This means
that we can set relatively low memory.low thresholds for variable or
bursty workloads while still getting a reasonable level of protection,
whereas with the previous version we may still trivially hit the 100%
clamp. The previous 100% clamp is also somewhat arbitrary, whereas this
one is more concretely based on the currently enforced protection
threshold, which is likely easier to reason about.
There is also a subtle issue with the way that proportional reclaim
worked previously -- it promotes having no memory.low, since it makes
pressure higher during low reclaim. This happens because we base our
scan pressure modulation on how far memory.current is between memory.min
and memory.low, but if memory.low is unset, we only use the overage
method. In most cromulent configurations, this then means that we end
up with *more* pressure than with no memory.low at all when we're in low
reclaim, which is not really very usable or expected.
With this patch, memory.low and memory.min affect reclaim pressure in a
more understandable and composable way. For example, from a user
standpoint, "protected" memory now remains untouchable from a reclaim
aggression standpoint, and users can also have more confidence that
bursty workloads will still receive some amount of guaranteed
protection.
Link: http://lkml.kernel.org/r/20190322160307.GA3316@chrisdown.name
Signed-off-by: Chris Down <chris@chrisdown.name>
Reviewed-by: Roman Gushchin <guro@fb.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Michal Hocko <mhocko@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Vladimir Davydov <vdavydov.dev@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Roman points out that when when we do the low reclaim pass, we scale the
reclaim pressure relative to position between 0 and the maximum
protection threshold.
However, if the maximum protection is based on memory.elow, and
memory.emin is above zero, this means we still may get binary behaviour
on second-pass low reclaim. This is because we scale starting at 0, not
starting at memory.emin, and since we don't scan at all below emin, we
end up with cliff behaviour.
This should be a fairly uncommon case since usually we don't go into the
second pass, but it makes sense to scale our low reclaim pressure
starting at emin.
You can test this by catting two large sparse files, one in a cgroup
with emin set to some moderate size compared to physical RAM, and
another cgroup without any emin. In both cgroups, set an elow larger
than 50% of physical RAM. The one with emin will have less page
scanning, as reclaim pressure is lower.
Rebase on top of and apply the same idea as what was applied to handle
cgroup_memory=disable properly for the original proportional patch
http://lkml.kernel.org/r/20190201045711.GA18302@chrisdown.name ("mm,
memcg: Handle cgroup_disable=memory when getting memcg protection").
Link: http://lkml.kernel.org/r/20190201051810.GA18895@chrisdown.name
Signed-off-by: Chris Down <chris@chrisdown.name>
Suggested-by: Roman Gushchin <guro@fb.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Dennis Zhou <dennis@kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
cgroup v2 introduces two memory protection thresholds: memory.low
(best-effort) and memory.min (hard protection). While they generally do
what they say on the tin, there is a limitation in their implementation
that makes them difficult to use effectively: that cliff behaviour often
manifests when they become eligible for reclaim. This patch implements
more intuitive and usable behaviour, where we gradually mount more
reclaim pressure as cgroups further and further exceed their protection
thresholds.
This cliff edge behaviour happens because we only choose whether or not
to reclaim based on whether the memcg is within its protection limits
(see the use of mem_cgroup_protected in shrink_node), but we don't vary
our reclaim behaviour based on this information. Imagine the following
timeline, with the numbers the lruvec size in this zone:
1. memory.low=1000000, memory.current=999999. 0 pages may be scanned.
2. memory.low=1000000, memory.current=1000000. 0 pages may be scanned.
3. memory.low=1000000, memory.current=1000001. 1000001* pages may be
scanned. (?!)
* Of course, we won't usually scan all available pages in the zone even
without this patch because of scan control priority, over-reclaim
protection, etc. However, as shown by the tests at the end, these
techniques don't sufficiently throttle such an extreme change in input,
so cliff-like behaviour isn't really averted by their existence alone.
Here's an example of how this plays out in practice. At Facebook, we are
trying to protect various workloads from "system" software, like
configuration management tools, metric collectors, etc (see this[0] case
study). In order to find a suitable memory.low value, we start by
determining the expected memory range within which the workload will be
comfortable operating. This isn't an exact science -- memory usage deemed
"comfortable" will vary over time due to user behaviour, differences in
composition of work, etc, etc. As such we need to ballpark memory.low,
but doing this is currently problematic:
1. If we end up setting it too low for the workload, it won't have
*any* effect (see discussion above). The group will receive the full
weight of reclaim and won't have any priority while competing with the
less important system software, as if we had no memory.low configured
at all.
2. Because of this behaviour, we end up erring on the side of setting
it too high, such that the comfort range is reliably covered. However,
protected memory is completely unavailable to the rest of the system,
so we might cause undue memory and IO pressure there when we *know* we
have some elasticity in the workload.
3. Even if we get the value totally right, smack in the middle of the
comfort zone, we get extreme jumps between no pressure and full
pressure that cause unpredictable pressure spikes in the workload due
to the current binary reclaim behaviour.
With this patch, we can set it to our ballpark estimation without too much
worry. Any undesirable behaviour, such as too much or too little reclaim
pressure on the workload or system will be proportional to how far our
estimation is off. This means we can set memory.low much more
conservatively and thus waste less resources *without* the risk of the
workload falling off a cliff if we overshoot.
As a more abstract technical description, this unintuitive behaviour
results in having to give high-priority workloads a large protection
buffer on top of their expected usage to function reliably, as otherwise
we have abrupt periods of dramatically increased memory pressure which
hamper performance. Having to set these thresholds so high wastes
resources and generally works against the principle of work conservation.
In addition, having proportional memory reclaim behaviour has other
benefits. Most notably, before this patch it's basically mandatory to set
memory.low to a higher than desirable value because otherwise as soon as
you exceed memory.low, all protection is lost, and all pages are eligible
to scan again. By contrast, having a gradual ramp in reclaim pressure
means that you now still get some protection when thresholds are exceeded,
which means that one can now be more comfortable setting memory.low to
lower values without worrying that all protection will be lost. This is
important because workingset size is really hard to know exactly,
especially with variable workloads, so at least getting *some* protection
if your workingset size grows larger than you expect increases user
confidence in setting memory.low without a huge buffer on top being
needed.
Thanks a lot to Johannes Weiner and Tejun Heo for their advice and
assistance in thinking about how to make this work better.
In testing these changes, I intended to verify that:
1. Changes in page scanning become gradual and proportional instead of
binary.
To test this, I experimented stepping further and further down
memory.low protection on a workload that floats around 19G workingset
when under memory.low protection, watching page scan rates for the
workload cgroup:
+------------+-----------------+--------------------+--------------+
| memory.low | test (pgscan/s) | control (pgscan/s) | % of control |
+------------+-----------------+--------------------+--------------+
| 21G | 0 | 0 | N/A |
| 17G | 867 | 3799 | 23% |
| 12G | 1203 | 3543 | 34% |
| 8G | 2534 | 3979 | 64% |
| 4G | 3980 | 4147 | 96% |
| 0 | 3799 | 3980 | 95% |
+------------+-----------------+--------------------+--------------+
As you can see, the test kernel (with a kernel containing this
patch) ramps up page scanning significantly more gradually than the
control kernel (without this patch).
2. More gradual ramp up in reclaim aggression doesn't result in
premature OOMs.
To test this, I wrote a script that slowly increments the number of
pages held by stress(1)'s --vm-keep mode until a production system
entered severe overall memory contention. This script runs in a highly
protected slice taking up the majority of available system memory.
Watching vmstat revealed that page scanning continued essentially
nominally between test and control, without causing forward reclaim
progress to become arrested.
[0]: https://facebookmicrosites.github.io/cgroup2/docs/overview.html#case-study-the-fbtax2-project
[akpm@linux-foundation.org: reflow block comments to fit in 80 cols]
[chris@chrisdown.name: handle cgroup_disable=memory when getting memcg protection]
Link: http://lkml.kernel.org/r/20190201045711.GA18302@chrisdown.name
Link: http://lkml.kernel.org/r/20190124014455.GA6396@chrisdown.name
Signed-off-by: Chris Down <chris@chrisdown.name>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Reviewed-by: Roman Gushchin <guro@fb.com>
Cc: Michal Hocko <mhocko@kernel.org>
Cc: Tejun Heo <tj@kernel.org>
Cc: Dennis Zhou <dennis@kernel.org>
Cc: Tetsuo Handa <penguin-kernel@i-love.sakura.ne.jp>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
The "mode" and "level" variables are enums and in this context GCC will
treat them as unsigned ints so the error handling is never triggered.
I also removed the bogus initializer because it isn't required any more
and it's sort of confusing.
[akpm@linux-foundation.org: reduce implicit and explicit typecasting]
[akpm@linux-foundation.org: fix return value, add comment, per Matthew]
Link: http://lkml.kernel.org/r/20190925110449.GO3264@mwanda
Fixes: 3cadfa2b94 ("mm/vmpressure.c: convert to use match_string() helper")
Signed-off-by: Dan Carpenter <dan.carpenter@oracle.com>
Reviewed-by: Andy Shevchenko <andriy.shevchenko@linux.intel.com>
Acked-by: David Rientjes <rientjes@google.com>
Reviewed-by: Matthew Wilcox <willy@infradead.org>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Thomas Gleixner <tglx@linutronix.de>
Cc: Enrico Weigelt <info@metux.net>
Cc: Kate Stewart <kstewart@linuxfoundation.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
There's a really hard to reproduce race in z3fold between z3fold_free()
and z3fold_reclaim_page(). z3fold_reclaim_page() can claim the page
after z3fold_free() has checked if the page was claimed and
z3fold_free() will then schedule this page for compaction which may in
turn lead to random page faults (since that page would have been
reclaimed by then).
Fix that by claiming page in the beginning of z3fold_free() and not
forgetting to clear the claim in the end.
[vitalywool@gmail.com: v2]
Link: http://lkml.kernel.org/r/20190928113456.152742cf@bigdell
Link: http://lkml.kernel.org/r/20190926104844.4f0c6efa1366b8f5741eaba9@gmail.com
Signed-off-by: Vitaly Wool <vitalywool@gmail.com>
Reported-by: Markus Linnala <markus.linnala@gmail.com>
Cc: Dan Streetman <ddstreet@ieee.org>
Cc: Vlastimil Babka <vbabka@suse.cz>
Cc: Henry Burns <henrywolfeburns@gmail.com>
Cc: Shakeel Butt <shakeelb@google.com>
Cc: Markus Linnala <markus.linnala@gmail.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Partially revert 16db3d3f11 ("kernel/sysctl.c: threads-max observe
limits") because the patch is causing a regression to any workload which
needs to override the auto-tuning of the limit provided by kernel.
set_max_threads is implementing a boot time guesstimate to provide a
sensible limit of the concurrently running threads so that runaways will
not deplete all the memory. This is a good thing in general but there
are workloads which might need to increase this limit for an application
to run (reportedly WebSpher MQ is affected) and that is simply not
possible after the mentioned change. It is also very dubious to
override an admin decision by an estimation that doesn't have any direct
relation to correctness of the kernel operation.
Fix this by dropping set_max_threads from sysctl_max_threads so any
value is accepted as long as it fits into MAX_THREADS which is important
to check because allowing more threads could break internal robust futex
restriction. While at it, do not use MIN_THREADS as the lower boundary
because it is also only a heuristic for automatic estimation and admin
might have a good reason to stop new threads to be created even when
below this limit.
This became more severe when we switched x86 from 4k to 8k kernel
stacks. Starting since 6538b8ea88 ("x86_64: expand kernel stack to
16K") (3.16) we use THREAD_SIZE_ORDER = 2 and that halved the auto-tuned
value.
In the particular case
3.12
kernel.threads-max = 515561
4.4
kernel.threads-max = 200000
Neither of the two values is really insane on 32GB machine.
I am not sure we want/need to tune the max_thread value further. If
anything the tuning should be removed altogether if proven not useful in
general. But we definitely need a way to override this auto-tuning.
Link: http://lkml.kernel.org/r/20190922065801.GB18814@dhcp22.suse.cz
Fixes: 16db3d3f11 ("kernel/sysctl.c: threads-max observe limits")
Signed-off-by: Michal Hocko <mhocko@suse.com>
Reviewed-by: "Eric W. Biederman" <ebiederm@xmission.com>
Cc: Heinrich Schuchardt <xypron.glpk@gmx.de>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In kdump kernel, memcg usually is disabled with 'cgroup_disable=memory'
for saving memory. Now kdump kernel will always panic when dump vmcore
to local disk:
BUG: kernel NULL pointer dereference, address: 0000000000000ab8
Oops: 0000 [#1] SMP NOPTI
CPU: 0 PID: 598 Comm: makedumpfile Not tainted 5.3.0+ #26
Hardware name: HPE ProLiant DL385 Gen10/ProLiant DL385 Gen10, BIOS A40 10/02/2018
RIP: 0010:mem_cgroup_track_foreign_dirty_slowpath+0x38/0x140
Call Trace:
__set_page_dirty+0x52/0xc0
iomap_set_page_dirty+0x50/0x90
iomap_write_end+0x6e/0x270
iomap_write_actor+0xce/0x170
iomap_apply+0xba/0x11e
iomap_file_buffered_write+0x62/0x90
xfs_file_buffered_aio_write+0xca/0x320 [xfs]
new_sync_write+0x12d/0x1d0
vfs_write+0xa5/0x1a0
ksys_write+0x59/0xd0
do_syscall_64+0x59/0x1e0
entry_SYSCALL_64_after_hwframe+0x44/0xa9
And this will corrupt the 1st kernel too with 'cgroup_disable=memory'.
Via the trace and with debugging, it is pointing to commit 97b27821b4
("writeback, memcg: Implement foreign dirty flushing") which introduced
this regression. Disabling memcg causes the null pointer dereference at
uninitialized data in function mem_cgroup_track_foreign_dirty_slowpath().
Fix it by returning directly if memcg is disabled, but not trying to
record the foreign writebacks with dirty pages.
Link: http://lkml.kernel.org/r/20190924141928.GD31919@MiWiFi-R3L-srv
Fixes: 97b27821b4 ("writeback, memcg: Implement foreign dirty flushing")
Signed-off-by: Baoquan He <bhe@redhat.com>
Acked-by: Michal Hocko <mhocko@suse.com>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Jan Kara <jack@suse.cz>
Cc: Tejun Heo <tj@kernel.org>
Cc: Jens Axboe <axboe@kernel.dk>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
We get two warnings when build kernel W=1:
mm/shuffle.c:36:12: warning: no previous prototype for `shuffle_show' [-Wmissing-prototypes]
mm/sparse.c:220:6: warning: no previous prototype for `subsection_mask_set' [-Wmissing-prototypes]
Make the functions static to fix this.
Link: http://lkml.kernel.org/r/1566978161-7293-1-git-send-email-wang.yi59@zte.com.cn
Signed-off-by: Yi Wang <wang.yi59@zte.com.cn>
Reviewed-by: David Hildenbrand <david@redhat.com>
Reviewed-by: Kees Cook <keescook@chromium.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
SECTION_SIZE and SECTION_MASK macros are not getting used anymore. But
they do conflict with existing definitions on arm64 platform causing
following warning during build. Lets drop these unused macros.
mm/memremap.c:16: warning: "SECTION_MASK" redefined
#define SECTION_MASK ~((1UL << PA_SECTION_SHIFT) - 1)
arch/arm64/include/asm/pgtable-hwdef.h:79: note: this is the location of the previous definition
#define SECTION_MASK (~(SECTION_SIZE-1))
mm/memremap.c:17: warning: "SECTION_SIZE" redefined
#define SECTION_SIZE (1UL << PA_SECTION_SHIFT)
arch/arm64/include/asm/pgtable-hwdef.h:78: note: this is the location of the previous definition
#define SECTION_SIZE (_AC(1, UL) << SECTION_SHIFT)
Link: http://lkml.kernel.org/r/1569312010-31313-1-git-send-email-anshuman.khandual@arm.com
Signed-off-by: Anshuman Khandual <anshuman.khandual@arm.com>
Reported-by: kbuild test robot <lkp@intel.com>
Reviewed-by: David Hildenbrand <david@redhat.com>
Cc: Dan Williams <dan.j.williams@intel.com>
Cc: Jason Gunthorpe <jgg@ziepe.ca>
Cc: Logan Gunthorpe <logang@deltatee.com>
Cc: Ira Weiny <ira.weiny@intel.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Calling 'panic()' on a kernel with CONFIG_PREEMPT=y can leave the
calling CPU in an infinite loop, but with interrupts and preemption
enabled. From this state, userspace can continue to be scheduled,
despite the system being "dead" as far as the kernel is concerned.
This is easily reproducible on arm64 when booting with "nosmp" on the
command line; a couple of shell scripts print out a periodic "Ping"
message whilst another triggers a crash by writing to
/proc/sysrq-trigger:
| sysrq: Trigger a crash
| Kernel panic - not syncing: sysrq triggered crash
| CPU: 0 PID: 1 Comm: init Not tainted 5.2.15 #1
| Hardware name: linux,dummy-virt (DT)
| Call trace:
| dump_backtrace+0x0/0x148
| show_stack+0x14/0x20
| dump_stack+0xa0/0xc4
| panic+0x140/0x32c
| sysrq_handle_reboot+0x0/0x20
| __handle_sysrq+0x124/0x190
| write_sysrq_trigger+0x64/0x88
| proc_reg_write+0x60/0xa8
| __vfs_write+0x18/0x40
| vfs_write+0xa4/0x1b8
| ksys_write+0x64/0xf0
| __arm64_sys_write+0x14/0x20
| el0_svc_common.constprop.0+0xb0/0x168
| el0_svc_handler+0x28/0x78
| el0_svc+0x8/0xc
| Kernel Offset: disabled
| CPU features: 0x0002,24002004
| Memory Limit: none
| ---[ end Kernel panic - not syncing: sysrq triggered crash ]---
| Ping 2!
| Ping 1!
| Ping 1!
| Ping 2!
The issue can also be triggered on x86 kernels if CONFIG_SMP=n,
otherwise local interrupts are disabled in 'smp_send_stop()'.
Disable preemption in 'panic()' before re-enabling interrupts.
Link: http://lkml.kernel.org/r/20191002123538.22609-1-will@kernel.org
Link: https://lore.kernel.org/r/BX1W47JXPMR8.58IYW53H6M5N@dragonstone
Signed-off-by: Will Deacon <will@kernel.org>
Reported-by: Xogium <contact@xogium.me>
Reviewed-by: Kees Cook <keescook@chromium.org>
Cc: Russell King <linux@armlinux.org.uk>
Cc: Greg Kroah-Hartman <gregkh@linuxfoundation.org>
Cc: Ingo Molnar <mingo@redhat.com>
Cc: Petr Mladek <pmladek@suse.com>
Cc: Feng Tang <feng.tang@intel.com>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In ocfs2_info_scan_inode_alloc(), there is an if statement on line 283
to check whether inode_alloc is NULL:
if (inode_alloc)
When inode_alloc is NULL, it is used on line 287:
ocfs2_inode_lock(inode_alloc, &bh, 0);
ocfs2_inode_lock_full_nested(inode, ...)
struct ocfs2_super *osb = OCFS2_SB(inode->i_sb);
Thus, a possible null-pointer dereference may occur.
To fix this bug, inode_alloc is checked on line 286.
This bug is found by a static analysis tool STCheck written by us.
Link: http://lkml.kernel.org/r/20190726033717.32359-1-baijiaju1990@gmail.com
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Gang He <ghe@suse.com>
Cc: Jun Piao <piaojun@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In ocfs2_write_end_nolock(), there are an if statement on lines 1976,
2047 and 2058, to check whether handle is NULL:
if (handle)
When handle is NULL, it is used on line 2045:
ocfs2_update_inode_fsync_trans(handle, inode, 1);
oi->i_sync_tid = handle->h_transaction->t_tid;
Thus, a possible null-pointer dereference may occur.
To fix this bug, handle is checked before calling
ocfs2_update_inode_fsync_trans().
This bug is found by a static analysis tool STCheck written by us.
Link: http://lkml.kernel.org/r/20190726033705.32307-1-baijiaju1990@gmail.com
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Gang He <ghe@suse.com>
Cc: Jun Piao <piaojun@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In ocfs2_xa_prepare_entry(), there is an if statement on line 2136 to
check whether loc->xl_entry is NULL:
if (loc->xl_entry)
When loc->xl_entry is NULL, it is used on line 2158:
ocfs2_xa_add_entry(loc, name_hash);
loc->xl_entry->xe_name_hash = cpu_to_le32(name_hash);
loc->xl_entry->xe_name_offset = cpu_to_le16(loc->xl_size);
and line 2164:
ocfs2_xa_add_namevalue(loc, xi);
loc->xl_entry->xe_value_size = cpu_to_le64(xi->xi_value_len);
loc->xl_entry->xe_name_len = xi->xi_name_len;
Thus, possible null-pointer dereferences may occur.
To fix these bugs, if loc-xl_entry is NULL, ocfs2_xa_prepare_entry()
abnormally returns with -EINVAL.
These bugs are found by a static analysis tool STCheck written by us.
[akpm@linux-foundation.org: remove now-unused ocfs2_xa_add_entry()]
Link: http://lkml.kernel.org/r/20190726101447.9153-1-baijiaju1990@gmail.com
Signed-off-by: Jia-Ju Bai <baijiaju1990@gmail.com>
Reviewed-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Changwei Ge <gechangwei@live.cn>
Cc: Gang He <ghe@suse.com>
Cc: Jun Piao <piaojun@huawei.com>
Cc: Stephen Rothwell <sfr@canb.auug.org.au>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
Unused portion of a part-written fs-block-sized block is not set to zero
in unaligned append direct write.This can lead to serious data
inconsistencies.
Ocfs2 manage disk with cluster size(for example, 1M), part-written in
one cluster will change the cluster state from UN-WRITTEN to WRITTEN,
VFS(function dio_zero_block) doesn't do the cleaning because bh's state
is not set to NEW in function ocfs2_dio_wr_get_block when we write a
WRITTEN cluster. For example, the cluster size is 1M, file size is 8k
and we direct write from 14k to 15k, then 12k~14k and 15k~16k will
contain dirty data.
We have to deal with two cases:
1.The starting position of direct write is outside the file.
2.The starting position of direct write is located in the file.
We need set bh's state to NEW in the first case. In the second case, we
need mapped twice because bh's state of area out file should be set to
NEW while area in file not.
[akpm@linux-foundation.org: coding style fixes]
Link: http://lkml.kernel.org/r/5292e287-8f1a-fd4a-1a14-661e555e0bed@huawei.com
Signed-off-by: Jia Guo <guojia12@huawei.com>
Reviewed-by: Yiwen Jiang <jiangyiwen@huawei.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>
Cc: Joseph Qi <joseph.qi@huawei.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
In commit 9f79b78ef7 ("Convert filldir[64]() from __put_user() to
unsafe_put_user()") I made filldir() use unsafe_put_user(), which
improves code generation on x86 enormously.
But because we didn't have a "unsafe_copy_to_user()", the dirent name
copy was also done by hand with unsafe_put_user() in a loop, and it
turns out that a lot of other architectures didn't like that, because
unlike x86, they have various alignment issues.
Most non-x86 architectures trap and fix it up, and some (like xtensa)
will just fail unaligned put_user() accesses unconditionally. Which
makes that "copy using put_user() in a loop" not work for them at all.
I could make that code do explicit alignment etc, but the architectures
that don't like unaligned accesses also don't really use the fancy
"user_access_begin/end()" model, so they might just use the regular old
__copy_to_user() interface.
So this commit takes that looping implementation, turns it into the x86
version of "unsafe_copy_to_user()", and makes other architectures
implement the unsafe copy version as __copy_to_user() (the same way they
do for the other unsafe_xyz() accessor functions).
Note that it only does this for the copying _to_ user space, and we
still don't have a unsafe version of copy_from_user().
That's partly because we have no current users of it, but also partly
because the copy_from_user() case is slightly different and cannot
efficiently be implemented in terms of a unsafe_get_user() loop (because
gcc can't do asm goto with outputs).
It would be trivial to do this using "rep movsb", which would work
really nicely on newer x86 cores, but really badly on some older ones.
Al Viro is looking at cleaning up all our user copy routines to make
this all a non-issue, but for now we have this simple-but-stupid version
for x86 that works fine for the dirent name copy case because those
names are short strings and we simply don't need anything fancier.
Fixes: 9f79b78ef7 ("Convert filldir[64]() from __put_user() to unsafe_put_user()")
Reported-by: Guenter Roeck <linux@roeck-us.net>
Reported-and-tested-by: Tony Luck <tony.luck@intel.com>
Cc: Al Viro <viro@zeniv.linux.org.uk>
Cc: Max Filippov <jcmvbkbc@gmail.com>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
